ADB Economics
Working Paper Series
Winners and Losers of Multinational Firm Entry
into Developing Countries:
Evidence from the Special Economic Zones
of the People’s Republic of China
Avraham Ebenstein
No. 276 | October 2011
ADB Economics Working Paper Series No. 276
Winners and Losers of Multinational Firm
Entry into Developing Countries:
Evidence from the Special Economic Zones
of the People’s Republic of China
Avraham Ebenstein
October 2011
Avraham Ebenstein is Professor of Economics in the Hebrew University of Jerusalem. The author
acknowledges generous financial support from the Asian Development Bank. Excellent research assistance
was provided by Revital Bar, Joan Fang, Eyal Frank, and Michael Freedman. Special thanks for comments
and helpful advice are due Yaohoui Zhao and Chuanchuan Zhang. The author accepts responsibility for
any errors in the paper.
Asian Development Bank
6 ADB Avenue, Mandaluyong City
1550 Metro Manila, Philippines
www.adb.org/economics
©2011 by Asian Development Bank
October 2011
ISSN 1655-5252
Publication Stock No. WPS114146
The views expressed in this paper
are those of the author(s) and do not
necessarily reflect the views or policies
of the Asian Development Bank.
The ADB Economics Working Paper Series is a forum for stimulating discussion and
eliciting feedback on ongoing and recently completed research and policy studies
undertaken by the Asian Development Bank (ADB) staff, consultants, or resource
persons. The series deals with key economic and development problems, particularly
those facing the Asia and Pacific region; as well as conceptual, analytical, or
methodological issues relating to project/program economic analysis, and statistical data
and measurement. The series aims to enhance the knowledge on Asia’s development
and policy challenges; strengthen analytical rigor and quality of ADB’s country partnership
strategies, and its subregional and country operations; and improve the quality and
availability of statistical data and development indicators for monitoring development
effectiveness.
The ADB Economics Working Paper Series is a quick-disseminating, informal publication
whose titles could subsequently be revised for publication as articles in professional
journals or chapters in books. The series is maintained by the Economics and Research
Department.
Contents
Abstract
v
I. Introduction
1
II. Background
2
A.
B.
2
4
III. Data
5
IV. Estimation Strategy 9
A.
B.
C.
V. Empirical Results
11
A.
B. C. 11
12
16
VI. Conclusion
21
Appendix
23
References
25
The PRC’s Special Economic Zones
Theoretical Predictions
Firm Entry and Average Productivity in Special Economic Zones
Firm Profits and Worker Wages in Special Economic Zones
Returns to Skill, Wage Inequality, and Special Economic Zones
Firm Entry Productivity, Profits, and Aggregate Wages
Wage Effects and Responses to Inequality 9
10
10
Abstract
This paper examines the impact of multinational firm entry into local labor
markets on employment, productivity, and wages. It exploits a natural experiment
associated with the People’s Republic of China’s rapid economic reforms and
assignment of cities to special economic zone status in the 1980s and 1990s.
Using data on both firms and workers, it is found that these policies increased
foreign direct investment, which raised average labor productivity in these labor
markets. However, only modest increases in median wage rates across these
cities are observed in the face of large increases in wage inequality and rising
local prices, limiting the benefits to most workers in these cities. Evidence is
presented that corporate profits captured most of the increase in productivity in
these areas.
I. Introduction
In 1978, the People’s Republic of China’s (PRC) premier Deng Xiao Ping launched the
country’s efforts to open its doors to the world. Since the changes that he envisioned
were drastic, he initiated a policy by which key strategic cities would be chosen as
experimental zones with privileged status. The special economic zones (SEZs) and
closely related free trade zones (FTZs) were spectacularly successful and attracted
foreign direct investment (FDI), as well as agglomeration of domestic firms hoping to do
business with multinational corporations. While the strategies used to attract investment
were extremely successful, little is known as to the success of the policy’s intention of
foreign investment inducing productivity improvement in domestic PRC firms. It is even
less clear whether the wage increases due to foreign investment has translated into wage
increases for Chinese workers, especially those with low skill levels.
Has the PRC’s phenomenal growth created a tide that “lifted all ships”? Many fear
that corporate profits have captured the bulk of the surplus generated by lower costs
of production in the PRC, leaving labor with a smaller share of an albeit larger pie. If
wage increases for workers are modest, rising average price levels may leave many
workers worse off as a result of trade, in terms of purchasing power. Examining the
PRC experience in these tax-privileged zones is important for evaluating the potential
of globalization to reduce poverty in developing countries, as the PRC experience has
been hailed as a model. It has been used to encourage other countries to pursue similar
development strategies, although in some countries such as India, local groups have
blocked the implementation of SEZs due to concern about their impact on the local
economy. Thus, an accurate assessment of the economic impact of these policies on
such factors as employment, wages, and income inequality is of significance both in the
PRC and in developing countries around the world.
This study exploits the phased rollout of the PRC’s SEZs, and closely related FTZs1,
to assess the impact of multinational activity on local labor markets, and the welfare of
workers in these cities. A key challenge in the analysis is that these areas were chosen
endogenously, as the majority of these areas were in coastal areas that may have
benefited from the PRC’s growth in the absence of special treatment. The study exploits a
quasi-experiment associated with Deng Xiao Ping’s famous “southern tour” in 1992. His
1
As will be discussed, there are several other special zones similar to SEZs and FTZs included in the analysis.
Specifically, export processing zones (EPZ) and coastal open cities (COC) are included in the analysis. An area is
generally referred to as a “special economic zone” when it has any of these four designations (SEZ, FTZ, EPZ, or
COC).
2 | ADB Economics Working Paper Series No. 276
visit and policy statements in favor of economic liberalization provided political impetus for
expanding the number of cities with special status, and increasing autonomy within the
existing SEZs.
This paper examines a panel of PRC cities—its firms, workers, and whether the cities
are tax-privileged each year. The sample period for firms (1950–2002) and for workers
(1988–2002) was characterized by large increases in FDI to the PRC, with much of the
increase occurring in the SEZs. Since the timing of the establishment of the zones varied
across cities, it is possible to estimate models that exploit variation within a city in terms
of FDI and other outcomes of interest such as labor productivity (proxied by valued added
per worker) and average wage levels.
The paper also examines is whether the location of higher productivity firms leads to
higher real wages for the average worker, for only the highest earning worker, or if the
extra productivity is completely captured by an increasing return to capital. The results
indicate large potential benefits and pitfalls to other Asian countries of rolling out similar
zones. These zones encourage foreign firm entry, and the agglomeration of domestic
firms that appear to benefit from technology spillovers. However, the potential benefit
of these policies is tempered by evidence of modest growth in real wages. It can be
concluded that the introduction of an SEZ in a context of cheap surplus labor may lead
to soaring corporate profits but be of little benefit to the average worker. While these
zones may lead to reallocation of labor from the rural to urban labor markets, the welfare
improvement for workers already in these cities is modest. Insofar as similar patterns of
rising wages and prices would follow from these zones in other Asian countries, social
insurance programs should be in place to protect low-skilled workers from experiencing
reduced purchasing power following these programs.
The paper is organized as follows. Section II discusses the background of how SEZs and
FTZs were established, and the special treatment they provided to multinational firms.
Section III presents data on firms, workers, and the rollout of special-status zones in the
PRC. Section IV presents the estimation strategy and Section V the empirical results.
Section VI concludes with a brief discussion of the policy implications of the findings.
II. Background
A.
The PRC’s Special Economic Zones
The SEZ experiment began soon after Deng Xiaoping’s 1978 policy statement in which
he argued for greater economic liberalization, and more interaction with firms from
overseas (Yeung et al. 2009). These zones were envisioned as small laboratories to
Winners and Losers of Multinational Firm Entry into Developing Countries:
Evidence from the Special Economic Zones of the People’s Republic of China | 3
explore the economic potential of a further opening up of the PRC’s economy, and so four
cities were chosen for SEZ status, in which they were able to operate with administrative
autonomy from the provincial government, and foreign firms were allowed tax exemptions.
The SEZs were strategically located in coastal areas close to islands with capitalist
economies, including Xiamen (near Taipei,China); Zhuhai (near Macao, China); and the
most successful SEZ, Shenzhen, which capitalized on its proximity to Hong Kong, China.
While the areas chosen were in locations convenient to foreign firms, they were by no
means already developed areas. In fact, the government focused on undeveloped cities
to minimize resistance to the new policies, and to limit damage should the experiment
fail. For example, prior to obtaining SEZ status in 1980, Shenzhen was a small fishing
village without even a single traffic light (CCPR 1987). Twenty years later, its population
had exceeded 10 million. The SEZs were successful at attracting foreign investment
and cheap migrant labor from nearby provinces almost immediately (Yeung et al. 2009).
By 1985, the SEZs accounted for more than 20% of the PRC’s FDI. The success of
the original SEZ cities spurred the government to open 14 coastal cities to foreign
investment in 1984 (Yeung and Hu 1992), which also began to attract additional foreign
firms attempting to capitalize on the PRC’s cheap labor and goods for lucrative overseas
consumer markets.
During the late 1980s, however, many policy makers in the PRC felt that the country’s
entry into the world economy was proceeding too slowly. Some believed that the PRC’s
reform efforts were stagnating and meant to develop faster, yet they met resistance from
conservative elements of the country who wished to maintain the status quo. Deng Xiao
Ping’s famous visit to the South in 1992 was intended to promote reform policies, and
embolden those who wished to continue the PRC’s move to capitalism. In the wake of
Deng’s visit, FTZs were established in several other coastal cities, including Shanghai’s
highly successful Putong Economic Zone. Smaller-scale economic and technological
development zones were also opened throughout the country, with many zones being
created in the country’s interior.
These zones have generated some controversy, however. Several scholars have
questioned the sustainability of SEZs due to a phenomenon known as “development
zone fever”: the opening of development zones beyond what is economically efficient or
feasible, which results in a waste of resources and arable land (Wong and Tang 2005,
Zhang 2011).2 Other scholars note that the preferential policy treatment accorded to
the SEZs are largely responsible for the sustained rise of regional disparity in the PRC,
particularly between the coastal provinces and the more Western provinces (Demurger
et al. 2002, Jones et al. 2003). According to one observer, the inequalities between
2
In 1991 there were only 117 development zones in the PRC, but by 1992 that had mushroomed to 2,700, the
overwhelming majority of which were independent initiatives by local authorities, from provincial governments
to township governments. In the mid-1990s the government began to clean up the development zone scene,
cancelling about 1,200 zones and returning more than 2 million mu of vacant land to agriculture (Deng and
Huang 2004).
4 | ADB Economics Working Paper Series No. 276
provinces in the PRC will lead to a rebellion as long and as arduous a struggle as the
Civil War in the United States (Tyler 1995).
B.
Theoretical Predictions
In the wake of globalization, a rich theoretical literature has developed to explain the
changes observed in both developed and developing countries. This section examines
how it may guide the analysis of the experience of the PRC’s SEZs. Trade theory that
examines the expected impact of the PRC’s entry into global markets often begin with
the standard insight of the Heckscher-Ohlin (H-O) model, which predicts that countries
will specialize in goods that require a resource-abundant input good. In the PRC, cheap
unskilled labor is a relatively abundant resource and developed economies, such as the
United States (US), have a greater supply of other resources, such as capital and highskilled labor. As predicted by the H-O model, in the wake of the PRC’s liberalized trade
policy, the US data reflect a decline in employment in industries that employ workers
who perform “routine” tasks, and a concurrent increase in imports in these industries
(Ebenstein et al. 2009). Insofar as routine tasks can be performed easily by foreign
workers, and are easily monitored by multinational parent corporations (Grossman and
Rossi-Hansberg 2009), it is logical to assume that SEZs will attract industries that require
large numbers of low-skilled workers, thereby increasing demand for their services.
These models would suggest that increased relative demand for low-skilled workers in
SEZs and other areas would raise the wages of unskilled workers, but they have more
ambiguous predictions for the impact on high-skilled workers. Antras et al. (2006) argue
that increased wage inequality in developing countries is a necessary implication of
offshoring. They consider a one-sector, two-country model in which large declines in
communications costs enable the formation of North–South teams. They argue that the
“globalization” equilibrium, wherein Northern workers can team up with Southern workers
at no additional expense, will lead to international teams in which Northern managers
supervise teams of Southern workers. They predict in their model that wage inequality
will be higher in the South following these changes: “Globalization improves the quality
of managers with whom Southern workers are matched, thus raising the productivity
of these workers, and thereby leading to an increase in the return to skill. This effect
is reinforced by an occupational choice effect: more agents become workers, hence
increasing the range of abilities in the worker distribution.” This prediction fits the PRC’s
SEZ experience, as its rural population left agriculture and entered into production jobs in
the SEZs. These models generate an unambiguous prediction that the return to skill will
rise.
Another strand of literature has focused on firm productivity, and the potential for highproductivity multinational firms to create technology spillovers to domestic firms in these
cities. In an influential paper, Melitz (2003) posits that only the most productive firms
will engage in multinational activity, and that trade can lead to welfare gains if lower-
Winners and Losers of Multinational Firm Entry into Developing Countries:
Evidence from the Special Economic Zones of the People’s Republic of China | 5
productivity firms are forced to exit the market. Indeed, empirical evidence suggests
that this claim is borne out by the data (Helpman et al. 2004). In combination with
evidence that domestic firms in the PRC are of lower productivity than American firms
through poor capital allocation (Hsieh and Klenow 2009), it may be that the SEZs create
an environment where domestic firms in the PRC are forced to either compete or exit
the market. The ensuing data analysis will examine how SEZ status has affected the
composition of firms, their productivity, and the wages of workers in these local labor
markets.
III. Data
The data on the timing and location of SEZs are the first attempt to catalogue these
events systematically.3 For each special zone, the year in which it was established, the
special privileges associated with the zone, and the county in which it is located are
recorded. The data covers six different types of zones: Special Economic Zones (SEZ),
Free Trade Zones (FTZ), Coastal Open Cities (COC), Economic Technology Development
Zone (ETDZ), Export Processing Zone (EPZ), and High Technology Development Zone
(HTDZ). The principal focus is on the SEZs, FTZs, EPZs, and COCs as these provided
much more substantial financial incentives to foreign multinational firms to enter the
market. As shown in Figure 1, there was a proliferation of these special-status areas
between 1985 and 1995, with a large number of special zones being established in the
early 1990s after Deng Xiao Ping’s visit to the South.
The firm data are taken from the 2003 Annual Survey of Industrial Production by the
National Bureau of Statistics. The sample is composed of all nonstate firms with more
than Yuan 5 million in revenue (about $800,000) plus all state-owned firms. The raw data
consist of over 100,000 firms. Importantly, for each firm both ownership type and year of
establishment are observed. As shown in Figure 2, foreign-firm entry rose precipitously in
the early 1990s, following the expansion of SEZs and expanded incentives within existing
zones. Summary statistics for the sample of firms are reported in Table 1.
3
These data are available for download at the author’s website, matched to the PRC’s 2000 census data at the
4-digit and 6-digit levels. See demog.berkeley.edu/~ebenstei/research/sez/datafiles/all_economic_zones_citygb.
dta.
6 | ADB Economics Working Paper Series No. 276
Figure 1: Geographic Distribution of Economic Zones
in the People’s Republic of China
Panel A: 1985
Panel B: 1995
Note: The figure displays all economic zones by prefecture in the listed year. The zones are special economic zones, free trade
zones, economic technology development zones, export processing zones, and high-tech development zones.
Source: Author’s calculations.
Winners and Losers of Multinational Firm Entry into Developing Countries:
Evidence from the Special Economic Zones of the People’s Republic of China | 7
Table 1: Summary Statistics for Sample of Firms in Manufacturing, 2003
Foreign Firm
(1=yes)
Number of Employees
(‘000s)
Total Annual Sales
(milion yuan)
Value Added
(million yuan)
Total Annual Profit
(million yuan)
Value Added per Worker
(thousand yuan)
Share of Annual Sales
Paid in Wages
Profit per Worker
(thousand yuan)
Special Economic Zone
(1=yes)
Established Prior to Economic Zones
(1950–1979)
Regular Special Economic Difference
City
Zone (ever)
(2)-(1)
(1)
(2)
(3)
Established After the Economic Zones
(1980–2002)
Regular Special Economic Difference
City
Zone (ever)
(5)-(4)
(4)
(5)
(6)
0.01
(0.10)
99.30
(784)
527.36
(1673)
28.95
(229.69)
3.10
(57.03)
41.43
(98.88)
0.134
(0.14)
1.554
(37.62)
0.00
0.02
(0.13)
113.24
(864)
460.54
(1717)
29.05
(212.81)
3.77
(57.02)
47.21
(83.08)
0.136
(0.14)
2.310
(28.19)
0.00
0.20
(0.40)
56.23
(365)
233.45
(605)
15.20
(94.46)
2.62
(29.71)
73.42
(244.11)
0.076
(0.08)
11.079
(66.01)
0.00
14,937
4,064
Observations
0.00715*
(0.004)
13.94
(19.44)
−66.82
(63)
0.10
(5.06)
0.67
(1.20)
5.78
(4.90)
0.002
(0.01)
0.756
(1.75)
0.00
105,957
0.37
(0.48)
89.03
(766)
245.65
(639)
23.39
(229.57)
5.33
(90.41)
93.26
(246.66)
0.084
(0.09)
17.065
(90.78)
0.69
(0.46)
42,567
0.178***
(0.04)
32.80***
(9.93)
12.21
(23)
8.187***
(2.65)
2.709***
(0.82)
19.84***
(6.60)
0.00774**
(0.00)
5.986***
(1.98)
0.691***
(0.09)
Note: The sample is composed of all manufacturing firms in 2003 that have annual revenue greater than RMB5 million ($800,000).
The designation of a city as a special economic zone also includes cities that are free trade zones, export processing zones,
or coastal open cities. The firms in columns 1 and 4 are located in cities without a special zone. Firms in columns 2 and 5 are
located in cities with special economic status, or cities that would eventually be given special economic status. Statistics are
reported using 2003 firm data, stratifying them by the year of opening.
Source: China Annual Survey of Manufacturing Firms (2003).
Figure 2: Foreign Firm Openings in the People’s Republic of China, 1980–2000
2,500
2,000
1,500
1,000
500
0
1980
82
84
86
88
90
Free Trade Zones
Export Processing Zones
All Other Cities
92
94
96
98
2000
Special Economic Zones
Coastal Open Cities
Note: A vertical line is placed at 1992, the year in which Deng Xiao Ping visited the PRC’s special trade areas and initiated
additional autonomy and tax exemptions for foreign firms.
Source: China Annual Survey of Manufacturing Firms (2003).
8 | ADB Economics Working Paper Series No. 276
Data on workers and their wages are drawn from the Urban Household Surveys (1988–
2002). For each worker in the sample, age, sex, years of education, broad measures
of industry and occupation, city, and monthly wage income are observed. The sample
is restricted to those workers in the labor market: men aged 16–59 and women aged
16–55. Excluded from the sample are those who report being either retired or disabled.
These data are matched by city to the sample of firms, and to the catalogue of SEZ
status by county and year. Summary statistics are reported in Table 2. The analysis of
the causal effect of foreign entry on local labor markets is based on the assumption that
SEZ status assignment induces higher-productivity firms to locate in these areas, and this
is exogenous to the existing composition of workers in the city. As shown in Table 2, the
workers are quite similar in the two sets of cities—but workers in the SEZ cities are more
likely to be working in a year X city cell with a zone in place.
Table 2: Summary Statistics for Sample of Workers in the PRC, 1988–2001
Prior to Deng’s Southern Tour
(1988–1992)
Regular
City
(1)
Log Wages
Real Log Wages
Age
Male (1=yes)
Years of Education
Consumer Price Index
Value Added per Worker1
Special Economic Zone
(1=yes)
Observations
1Value
Special Economic Difference
Zone (ever)
(1)-(2)
(2)
(3)
7.44
7.60
(0.58)
(0.57)
7.00
7.14
(0.57)
(0.54)
36.67
37.18
(9.91)
(9.88)
0.52
0.160***
After Deng’s Southern Tour
(1993–2001)
Regular
City
(4)
Special Economic Difference
Zone (ever)
(4)-(5)
(5)
(6)
8.75
8.97
0.220*
(0.78)
(0.78)
(0.12)
7.76
7.83
0.07
(0.04)
(0.83)
(0.76)
(0.18)
0.511***
38.12
39.37
1.248***
(0.21)
(15.48)
(15.15)
(0.45)
0.52
0.00
0.51
0.51
–0.00
(0.50)
(0.50)
(0.00)
(0.50)
(0.50)
(0.01)
10.60
10.73
0.14
10.15
10.45
0.30
(2.69)
(2.63)
(0.10)
(3.83)
(3.56)
(0.29)
(0.05)
0.144***
156.46
158.83
2.38
288.32
317.24
28.92
(17.17)
(15.80)
(1.87)
(70.73)
(37.31)
(20.09)
41.31
65.02
66.55
77.55
11.01*
(19.73)
(17.09)
0.00
0.62
0.00
(0.49)
73,084
16,944
23.71***
(22.74)
(16.09)
0.619***
(3.85)
0.00
0.85
(0.10)
0.00
(0.36)
111,417
49,601
(6.51)
0.852***
(0.08)
added per worker is defined by taking the average among firms operating in the worker’s city in a particular year, using the
2003 firm-level data set.
Notes: The designation of a city as a special economic zone also includes cities that are free trade zones, export processing zones,
or coastal open cities. The workers and firms in columns 1 and 4 are located in cities without a special zone. Workers and
firms in columns 2 and 5 are located in cities with special economic status, or cities that would eventually be given special
economic status. The workers sample is restricted to individuals earning wage income among men aged 16–59 and women
aged 16–55.
Sources:China Urban Household Survey (1988–2001), China Annual Survey of Manufacturing Firms (2003).
Winners and Losers of Multinational Firm Entry into Developing Countries:
Evidence from the Special Economic Zones of the People’s Republic of China | 9
IV. Estimation Strategy
A.
Firm Entry and Average Productivity in Special Economic Zones
The proposed mechanism for workers to benefit from an economic zone is that foreign
firms are more productive on average, and their presence will induce increases in
average productivity, which may then lead to higher wages. Since special zones attract
foreign firms and higher-productivity domestic firms, of interest are the reduced form
relationship between SEZ status and labor productivity, as well as the mechanisms
behind a relationship. As a prelude, it is demonstrated that FDI and domestic firms (DOM)
are drawn to an SEZ in city j in year t, after accounting for city and year fixed effects.
Models can be estimated by regressions of the following form:
FDI jt = α + β1SEZ jt + µ j + µt + ε jt
(1)
DOM jt = α + β 2SEZ jt + µ j + µt + ε jt
(2)
Next, the relationship between average firm productivity and the status of the city as a
special zone is determined. Since the adoption of an SEZ may lead to foreign firm entry,
it is anticipated that less productive firms will leave the market, and spillovers between
foreign firms and domestic firms will lead to increased productivity. Labor productivity
averages across firms by city and year are calculated; these are considered to be the
primary mechanisms for an SEZ to attain better labor market outcomes. Since there are
no data on firms for the period by year, 2003 data, which report the year the firm opened,
are exploited. If firms in other areas “catch up”, this assumption will be conservative and
understate the productivity improvement from having new firms enter the SEZ. The use
of 2003 data for productivity measures by year is more plausible for foreign firms, which
presumably implement best practices from mature foreign markets. However, if domestic
firms in SEZs improve their productivity with a lag, labor productivity in any year may be
measured incorrectly with this strategy.
The valued added per employee of all firms i already opened in city j is calculated as
follows:
LPj =
i =I
VA
∑ EMPi
i =1
i
These data are then averaged among all firms opened by year t in city j. This allows
calculation of labor productivity and estimation of the reduced form relationship between
labor productivity in city j and year t, after accounting for year and city fixed effects.
10 | ADB Economics Working Paper Series No. 276
LPjt = α + β3SEZ jt + µ j + µt + ε ijt
B.
(3)
Firm Profits and Worker Wages in Special Economic Zones
Multinational firms, relative to their domestic counterparts, are more capital-intensive
and have higher productivity (Hsieh and Klenow 2009). In combination with the tax
advantages granted to these firms, their access to low-wage Chinese workers and highprice product markets in developed countries to increase the profits of firms in these
cities can be anticipated. The extent to which workers share the benefits is an open
empirical question. While standard microeconomic theory suggests that workers will earn
their marginal revenue product, and that SEZ status will lead to higher wages, many fear
that the surplus labor from rural areas will lead to firms holding an “upper hand” in wage
bargaining. Of interest here are both the impact on corporate profits and the impact on
workers' wages. Equations (4) and (5) estimate the reduced form relationship between
firm profits, workers' wages, and SEZ status.
Π jt = α + β 4SEZ jt + µ j + µt + ε jt
(4)
w jt = α + β 4SEZ jt + µ j + µt + ε jt (5)
Since equation (2) can be thought of as a first-stage equation for equation (5), there is
need to estimate models of the following form, where fitted values from equation (2) and
controls for individual characteristics X are used.

w ijt = α + β3LPjt + µ j + µt + X ijt + ε ijt
C.
(6)
Returns to Skill, Wage Inequality, and Special Economic Zones
A voluminous theoretical literature on how foreign firm entry affects skill prices exists,
but there is little real-world evidence. Here, models of the return to education within a
city-year cell, and how it relates to conversion of a city to SEZ status are considered.
To determine the impact on wage inequality, how individual-level wages are responsive
to the presence of an SEZ, and whether these have led to increases in wage inequality,
are examined. This includes impact of SEZ status on workers at the kth percentile of
wages. An alternative strategy is to consider how skill prices themselves have changed,
as proxied by years of education, in SEZ cities versus other cities. While the empirical
analysis will only examine these issues in a descriptive manner, the models are presented
for expository purposes.
w kjt = α + β 6SEZ jt + µ j + µt + ε jt
w ijt = α + β7SEZ jt ⋅ Yrsed ijt + µ j + µt + X ijt + ε jt
(7)
(8)
Winners and Losers of Multinational Firm Entry into Developing Countries:
Evidence from the Special Economic Zones of the People’s Republic of China | 11
V. Empirical Results
A.
Firm Entry
Table 3 estimates equations (1) and (2) to examine the link between firm entry and SEZs.
Focusing on SEZs, FTZs, EPZs, COCs and a variable, which is a combination of the
four different zones, regression is run on a city's number of firm openings for each type
of zone, with year and city fixed effects. The sample is the 345 distinct prefectures (or
cities) for the period 1951–2002, yielding 17,940 observations (345 x 52). Columns 1–3
differentiate between foreign firm openings, domestic firm openings, and state-owned firm
openings.
Table 3: Firm Entry in the PRC’s Economic Zones
Any Zone
Special Economic Zone
Free Trade Zone
Export Processing Zone
Coastal Open City
Foreign Firm
Openings
(1)
33.74***
(9.96)
31.20**
(12.93)
80.36***
(25.39)
14.64***
(4.83)
25.77***
(7.94)
Domestic Firm
Openings
(2)
21.56***
(7.37)
2.06
(2.66)
53.44**
(21.43)
22.15***
(7.21)
26.42**
(10.29)
State-Owned
Firm Openings
(3)
19.60***
(6.39)
2.11
(1.45)
36.27***
(13.93)
16.57***
(4.19)
18.43**
(7.70)
Note: N=17,940. Each observation is a city X year cell for the period 1951–2002 (345 x 52). Each cell in the table represents the
coefficient from a separate regression. In each regression, a city’s number of firm openings in a given year is regressed on
the listed zone status. Each regression has year and city fixed effects. Standard errors are clustered at the city level.
Source: China Annual Survey of Manufacturing Firms (2003).
The results show a positive and statistically significant relationship between all types of
firm entry and a city’s status as an SEZ. Interestingly, the different types of economic
zones had different effects on the entry of firms. As shown in row 1 of Table 2, the cities
assigned to SEZ status had a strong impact on foreign firm openings (21.41) but a very
small effect on domestic firm entry. As intended, these SEZ cities enabled foreign firms
to produce goods in the PRC for export—but did not generate a large number of new
domestic firms in these areas. In contrast, a city’s assignment to FTZ status increased
the number of foreign firm openings by 80.36 per year, and also resulted in an additional
53.44 new domestic private firms, and 36.27 state-owned firms. This is logical since
FTZs were established to enable trading of intermediate goods between foreign and
domestic firms. The results for EPZs and COCs are similar, with these areas generating a
statistically significant number of new domestic and foreign firms. 12 | ADB Economics Working Paper Series No. 276
These patterns are shown graphically in Figure 3, which tracks the expansion of firms
in the SEZs and all other cities. The zones were characterized by rapid increases in
the number of foreign firms, followed shortly thereafter by increases in the number of
domestic firms. In nonzone cities, domestic firm openings have been consistently higher
than foreign firm openings. As will be shown in the next section, foreign firms are on
average more productive, implying that these differences could generate differences in
profits and wages across cities.
Figure 3: Foreign and Domestic Firm Entry into the PRC’s Cities
Special Economic Zones
All Other Cities
3,000
10,000
Number of Openings
Number of Openings
8,000
2,000
1,000
6,000
4,000
2,000
0
1980
84
88
92
96
2000
Foreign Firms
0
1980
84
88
92
96
2000
Domestic Firms
Notes: A vertical line is placed at 1992, the year in which Deng Xiao Ping visited the PRC’s special trade areas and initiated
additional autonomy and tax exemptions for foreign firms.
Source: China Annual Survey of Manufacturing Firms (2003).
B. Productivity, Profits, and Aggregate Wages
Table 4 estimates equations (3), (4), and (5) to examine the link between economic
zones and several measures of economic performance for cities that are assigned
special status: total employment, value added per worker, profit per worker, and wages
per worker. These outcomes are regressed on a city’s zone status in a particular year,
including in all models year and city fixed effects. Unfortunately, there is no access to
data for the entire sample period (1951–2002) that would reliably account for economic
performance at the subnational level. Hence, the 2003 firm survey is used to construct a
Winners and Losers of Multinational Firm Entry into Developing Countries:
Evidence from the Special Economic Zones of the People’s Republic of China | 13
data set that has the year of opening, and which generates a distribution of firms by city
in each year. This heroically assumes that firms operate at their 2003 levels, with no exit
and entry. This is an attempt to analyze what types of firms entered cities before and after
the SEZ assignment, and in the absence of data for the period, constitutes a reasonable
proxy for the composition of firms by city and year during the period. While it is assumed
that firms neither exit nor enter, the data set is only for firms that are sufficiently large
(with greater than 5 million RMB in annual sales), and so this sample is presumably less
volatile than a sample aimed at small businesses. Also, while it is not ideal to assume
that firms operated at their current levels of productivity, this may be a conservative
assumption for the purpose of evaluating whether there are productivity differences
between domestic and foreign firms. If foreign firms provide technology spillovers to
domestic firms, assuming domestic firms operate at their 2003 levels will presumably
understate the true productivity differences between the sets of firms.
Table 4 uses this “pseudo-panel” data to estimate the reduced form relationship between
zone status and the aforementioned economic performance measures, with the models
estimated using both levels (columns 1–4) and logs (5-8) for the dependent variables.
It is worth noting that since this is imputed data, the magnitudes have no real “units”.
However, the analysis still can speak to whether zone status was associated with
differential firm performance. Column 1 indicates that there is a positive and significant
relationship between being an economic zone and total employment for all of the zones.
The data indicate that the FTZs generate the largest increase in employment (336,888)
and coastal open cities the smallest (130,101), and both estimates are statistically
significant at the 1% level. Columns 2 and 3 indicate similar increases in value added per
worker and profit per worker, which suggests that the firms entering in response to a city
having special status were more productive than incumbent firms. As shown in the table,
the existence of any zone is associated with a 59% increase in total employment and
a 74% profit per worker.4 However, as shown in column 4, there is no reported increase
in wages per worker in these cities. None of the zones experienced a large increase in
wages per worker, as reported by the firms themselves. This seems to indicate that the
main benefactors of the increased productivity from firm openings and economic zones
were the corporations, not the workers. The results indicate that the zones increased the
scale of industry and the number of jobs in manufacturing, but had very modest impacts
on worker compensation rates. It is worth noting, however, that the extra employment
may have allowed rural workers to relocate from farming into industry. If industry pays
higher wages, the results may understate the welfare benefit to the PRC of having SEZs.
However, these results do suggest that the assignment of zones did not translate into
higher wages for workers.
4
Note that the log specification will delete firms that had negative or zero profits.
Wages per
Worker
(4)
0.63
(0.71)
2.67
(2.88)
0.20
(0.71)
−0.45
(0.35)
−0.04
(0.60)
Total
Employment
(5)
0.587**
(0.28)
2.202**
(0.92)
0.500***
(0.19)
−0.17
(0.16)
0.30
(0.19)
LHS: Logs
Value Added
Profit per
per Worker
Worker
(6)
(7)
0.07
0.738***
(0.09)
(0.15)
0.394***
0.910***
(0.13)
(0.26)
−0.06
0.681**
(0.14)
(0.29)
−0.07
0.664***
(0.07)
(0.14)
−0.03
0.643***
(0.12)
(0.21)
Wages per
Worker
(8)
0.00
(0.06)
0.20
(0.21)
−0.05
(0.06)
−0.123***
(0.03)
−0.05
(0.06)
Note: N=17,940. Each observation is a city X year cell for the period 1951–2002 (345 x 52). Each cell in the table represents the coefficient from a separate regression. In each
regression, the listed zone status is regressed on a city’s aggregate employment, value added per worker among firms, or profit among firms. Each regression has year and city
fixed effects. Standard errors are clustered at the city level. The sample used in this table is constructed by creating a pseudo-panel of firm averages across cities and years by
exploiting the firm’s year of establishment and creating a set of firms thought to be operating in a given city and year. It is assumed that the firms operate at 2003 levels of
productivity, profit, employment, and wage levels in all years.
Source: China Annual Survey of Manufacturing Firms (2003).
Coastal Open City
Export Processing Zone
Free Trade Zone
Special Economic Zone
Any Zone
Total
Employment
(1)
170,359***
(40,779)
155,481*
(87,431)
336,888***
(101,340)
241,742***
(59,245)
143,404***
(45,095)
LHS: Levels
Value Added
Profit per
per Worker
Worker
(2)
(3)
17.65***
5.510***
(4.48)
(1.31)
33.61***
6.687***
(9.56)
(2.33)
11.11
6.529**
(7.10)
(3.18)
13.83***
6.213***
(4.35)
(1.51)
12.32**
4.113***
(5.58)
(1.44)
Table 4: Employment, Value Added, and Profit in Cities with Economic Zones
14 | ADB Economics Working Paper Series No. 276
Winners and Losers of Multinational Firm Entry into Developing Countries:
Evidence from the Special Economic Zones of the People’s Republic of China | 15
Figures 4 and 5 confirm graphically that firm productivity was higher in the economic
zones (SEZs, FTZs, EPZs, COCs) than in other cities. As shown in Figure 4, when firms
are stratified by the year in which they were established and by foreign or domestic
ownership, foreign firms are more productive than domestic firms, a difference persisting
throughout the sample period. However, the figure does indicate that the productivity
gap between the different types of firms is narrowing—which may indicate a technology
spillover between foreign firms and domestic/state-owned firms. In light of the productivity
difference between foreign versus domestic firms, Figure 5 presentes differences in
productivity between SEZs and other cities. Cities with SEZs experienced an average firm
productivity increase from Yuan 60,000 per worker in 1980 to Yuan 110,000 per worker in
2003, whereas growth in productivity was more sluggish in others cities, increasing from
Yuan 50,000 in 1980 to Yuan 80,000 per worker in 2003. These results suggest that a
widening gap has emerged between firms in the zones versus other parts of the country.
Whether this productivity improvement has induced an improvement in the real wage paid
to workers is examined further.
Figure 4: Trends in Productivity by Ownership Type
Yuan per Worker (’000s)
120
100
80
60
40
1960
65
70
75
80
85
90
95
2000
Year Firm Established
Foreign Private
Note:
Domestic Private
State-Owned
The sample is composed of all firms with greater than RMB500,000 in 2003. The graph is created using lowess smoother
with bandwidth .20 on annual averages of value-added per employee in special zones versus all other cities. The special
zones include special economic zones, free trade zones, export processing zones, and coastal open cities. Vertical lines
are placed to indicate (i) the establishment of special economic zones in 1980, and (ii) the expansion of zones and their
priveleges in 1992 following Deng Xiao Ping's visit to the South.
Source: China Annual Survey of Manufacturing Firms (2003).
16 | ADB Economics Working Paper Series No. 276
Figure 5: Trends in Productivity: Economic Zones and All Other Cities
Yuan per Worker (’000s)
120
100
80
60
40
1960
65
70
75
80
85
90
95
2000
Year Firm Established
Special Economic Zones
All Other Cities
Notes: The sample is composed of all firms with greater than RMB500,000 in 2003. The graph is created using lowess smoother
with bandwidth .20 on annual averages of value-added per employee in special zones versus all other cities. The special
zones include special economic zones, free trade zones, export processing zones, and coastal open cities. Vertical lines
are placed to indicate (i) the establishment of special economic zones in 1980, and (ii) the expansion of zones and their
priveleges in 1992 following Deng Xiao Ping's visit to the South.
Source: China Annual Survey of Manufacturing Firms (2003).
C. Wage Effects and Responses to Inequality
Table 5 estimates equations (4) and (5) to examine the link between average labor
productivity and wages, with a focus on how the productivity boost generated by SEZ
status leads to higher wages. Both ordinary least squares are and two-stage least
squares models are used to estimate these separately for all workers (column 1),
manufacturing workers (column 2), and nonmanufacturing workers (column 3).
All regressions include year fixed effects and province fixed effects.5 A rich set of
demographic controls including age, age squared, years of education, and sex, isolate the
impact of firm composition in the city on the compensation of workers.
5
While ideally, city fixed effects are included, data on workers before and after the rollout of zones are insufficient
to estimate stable models with city effects. The results should therefore be interpreted with caution, as city-specific
factors that affect wages cannot be accounted for.
Winners and Losers of Multinational Firm Entry into Developing Countries:
Evidence from the Special Economic Zones of the People’s Republic of China | 17
Table 5: OLS and 2SLS Models of the Impact of Average Labor Productivity on Wages Using
SEZ Status as an Instrument
Dependent Variable: Real Log Wages
Overall
Manufacturing
Other
(1)
(2)
(3)
Panel A: Reduced form OLS models of real log wages on a city’s SEZ status
City is an SEZ (1=yes)
0.09
0.08
0.09
(0.07)
(0.08)
(0.07)
R-squared
0.51
0.46
0.55
Panel B: OLS models of real log wages on log of average labor productivity
Log of average labor
0.203***
0.234***
0.195***
Productivity
(0.04)
(0.05)
(0.04)
R-squared
0.52
0.47
0.56
Panel C: OLS models of log of average labor productivity on SEZ status
City is an SEZ (1=yes)
0.235**
0.221**
0.243**
(0.09)
(0.09)
(0.10)
R-squared
0.58
0.59
0.58
Panel D: 2SLS models of real log wages on log of average labor productivity using SEZ status as the
instrumental variable
Log of average labor
0.37
0.38
0.370**
Productivity
(0.22)
(0.30)
(0.17)
R-squared
0.51
0.46
0.55
Demographic controls
Yes
Yes
Yes
Year fixed effects
Yes
Yes
Yes
Province fixed effects
Yes
Yes
Yes
Observations
137,320
51,540
85,780
OLS = ordinary least squares, SEZ = special economic zone, 2SLS = two-stage least squares.
Note: Each cell in the table represents the coefficient from a separate regression. Our designation of a city as a Special Economic
Zone also includes cities that are Free Trade Zones, Export Processing Zones, or Coastal Open Cities. Average labor
productivity is measured among all firms that were operating in the city in the year, assuming they were operating at their
2003 levels of productivity. The sample is restricted to individuals with wage income. Demographic controls for age, age
squared, years of education, and sex are suppressed. Standard errors are clustered at the city level.
Source: China Urban Household Survey (1988–2001), China Annual Survey of Manufacturing Firms (2003).
As shown in Panel A, workers in SEZs are paid somewhat better than workers in other
cities. These workers enjoy an 8–9% premium in real wages, but the difference is not
statistically significant. Upon examination of the relationship between wage rates and
labor productivity, as proxied by value added per worker among the manufacturing firms
in the sample, Panel B shows a robust relationship between wage rates and the log of
average productivity among firms in the city, with a 1% increase in average productivity
increasing real wage rates by 0.2%. This is statistically significant at the 1% level for
manufacturing and nonmanufacturing workers, suggesting a spillover into other sectors
of the economy. However, it is worth noting that this relationship may be endogenous.
If workers who are more productive sort into cities with more productive firms, it may
be that the causal link is more modest. Alternatively, if productivity is poorly measured
relative to wages, it may be that the relationship is even stronger than the findings. Panel
18 | ADB Economics Working Paper Series No. 276
C and D examines the link between average firm productivity and wage rates exploiting
the SEZ status of a city. Panel C reports the first-stage results, where a city’s average
productivity is regressed on its status as an SEZ. SEZ status increases productivity
by 23.5% among firms, significant at the 5% level. Using the fitted values from this
regression to examine the relationship between average productivity and real wages,
workers are rewarded with 37% of the increase in average labor productivity, leaving 63%
presumably to firms. However, these estimates are imprecise and statistically insignificant.
While the results are not definitive, they suggest that workers have received at most one
third of the increase in firm productivity in increased wages. In combination with evidence
from firms that SEZ status is associated with increased firm productivity and firm
profitability but is not associated with a higher share of revenue going to wages, it can be
concluded that firms have not passed on the increased productivity to their workers. Poor
bargaining power among workers due to large numbers of surplus labor in rural areas of
the PRC has led to a situation in which economic expansion leads to more jobs but not
necessarily better paying jobs within these cities in real terms.
Figure 6 indicates that one mechanism for tepid growth in real wages in the SEZ areas
was the rapid increase in prices. The figure indicates that from 1988 to 2001, the urban
consumer price index rose from a level of 100 in 1985 to 350 by 2001, a stunning 250%
increase. The price increase was slightly more modest in other cities, reaching 250 in
2001. The rapid increases in prices suggest that the workers and consumers were not
able to enjoy fully the fruits of the PRC’s expansion. In particular, if SEZs attract foreign
firms, and this leads to higher prices but no increases in the real wage, the welfare
impacts of these policies are ambiguous.
Figure 6: Trends in Prices: Special Economic Zones and All Other Cities
Urban CPI Index
(1985=100)
350
300
250
200
150
1988
90
92
94
Special Economic Zones
96
98
2000
All Other Cities
CPI = consumer price index.
Note: The sample is composed of all workers. The urban CPI is available by province and year, and assigned to the workers in the
sample. A vertical line is placed at 1992, the year in which Deng Xiao Ping visited the PRC’s special trade areas and initiated
additional autonomy and tax exemptions for foreign firms.
Sources:China Urban Consumer Price Index (1988–2001), China Urban Household Survey (1988–2001).
Winners and Losers of Multinational Firm Entry into Developing Countries:
Evidence from the Special Economic Zones of the People’s Republic of China | 19
This possibility is highlighted by Figure 7, which indicates that among the sample of
workers, little difference is observed in wage trends between SEZs and all other cities.
From 1988 to 2001, the pace of increase of worker's wages has remained similar in the
two groups, despite the productivity gap between firms in SEZs versus other cities. In
addition, the figure suggests a dramatic increase in inequality for both SEZs and other
cities.6 In 1988, while the gap between the 90th percentile and the 10th percentile for real
wages was less than Yuan 2000 (in 1985 units), the difference in 2002 was nearly Yuan
10,000. If the PRC’s economic boom is generating a windfall for firms but large increases
in inequality, it is unclear whether the growth will generate welfare improvements for
majority of the population.
Figure 7: Trends in Real Wages: Special Economic Zones and All Other Cities
Real Wages (Yuan)
10,000
8,000
6,000
4,000
2,000
0
1988
90
92
94
96
98
2000
Special Economic Zones (90th percentile)
Special Economic Zones (median)
Special Economic Zones (10th percentile)
All Other Cities (90th percentile)
All Other Cities (median)
All Other Cities (10th percentile)
Note:
The sample is composed of all workers. A vertical line is placed at 1992, the year in which Deng Xiao Ping visited the PRC’s
special trade areas and initiated additional autonomy and tax exemptions for foreign firms.
Source: China Urban Household Surveys (1988–2001).
The combination of rising inequality and anemic wage growth while corporate profits
have surged is a potential recipe for unrest or at least dissatisfaction. This possibility
is investigated in Table 6. The results indicate that workers in SEZs are more likely to
believe the city's income distribution is “very unfair”. Workers in SEZ cities are 2.67
percentage points more likely to characterize the city as such, which is a 10% increase
over a 30% sample mean (who report the city’s wage distribution is very unfair). Almost
no difference by gender is observed but a large gap by income and education is
6
Quantile regressions as an analogue to this figure are also estimated. The results indicate no significant gap
between SEZ cities or non-SEZ cities at various points in the wage distribution. The results are available upon
request.
20 | ADB Economics Working Paper Series No. 276
exhibited. A 10% increase in log wages is associated with a 6 percentage point decline in
the probability of perceiving the wage distribution as unfair, or a 20% decline. The table
also indicates that the SEZ status of a city is associated with the belief that the wage
distribution is unfair only among those with less than a high school degree. Being young
(under 40) and having more education are both negatively correlated with the perception
that a city's income distribution is unfair. This suggests that young and highly educated
workers perceive a fair return to education, and are therefore less likely to feel that the
city's income distribution is unfair.
Table 6: Perception of Fairness of Income Distribution, Chinese Workers (2002)
Sample Average
City is an SEZ (1=yes)
Sex
Log wage
Communist Party member
(1=yes)
Observations
R-squared
Dependent Variable:
Is the city’s income distribution
very unfair? (1=yes)
Overall
Less than HS HS Graduates
Education
(1)
(2)
(3)
0.0267**
0.0457***
0.000
(0.013)
(0.017)
(0.019)
0.0049
0.0087
0.0047
(0.012)
(0.017)
(0.018)
−0.0620***
−0.0511***
−0.0669***
(0.009)
(0.012)
(0.014)
−0.002
0.0316*
−0.022
(0.012)
(0.018)
(0.017)
6,581
3,722
2,859
0.013
0.011
0.013
HS = high school, SEZ = special economic zone..
Note: SEZs include cities that are free trade zones or export processing zones. The sample is restricted to household heads. Also
include are controls for age and age squared, which are suppressed. Standard errors are robust.
Source: China Household Income Survey (2002).
Figure 8 provides evidence in favor of this hypothesis. The figure indicates rapid
increases in the returns to education in the PRC, with the return rising from 3% to 4%
in 1988 to almost 14% in 2001. These trends are similar, however, in both SEZs and all
other cities. While it is not possible to identify noticeable gaps in the return to education
in the two areas, it may be that the PRC’s foreign firms have generated an increase in
the return to skill, but the effect is felt across the PRC rather than in a specific set of
labor markets. This topic could be an area for future work, though it is clear that rapid
increases in skill prices in combination with rising price levels could have important
welfare implications and lead to declining purchasing power among the PRC’s least
skilled workers.
Winners and Losers of Multinational Firm Entry into Developing Countries:
Evidence from the Special Economic Zones of the People’s Republic of China | 21
Figure 8: Trends in Returns to Education: Special Economic Zones
and All Other Cities (percent)
0.15
0.10
0.05
0
1988
90
92
94
96
98
2000
Returns to Education in Special Economic Zones
Returns to Education in Other Cities
Notes: The sample is composed of all workers. For each year, log wages are regressed on a worker's years of education, age, age
squared, and sex. The figure reports the coefficient on years of education among workers in special economic zones and
other cities. A vertical line is placed at 1992, the year in which Deng Xiao Ping visited the PRC’s special trade areas and
initiated additional autonomy and tax exemptions for foreign firms.
Source: China Urban Household Surveys (1988–2001).
VI. Conclusion
This paper examined the impact on both firms and workers of assigning a city special
economic zone status. While these areas have been hailed as a model for other Asian
countries looking to capitalize on globalization and access to foreign markets, the
Chinese experiment proves there are both winners and losers. Foreign multinationals
raise employment levels in manufacturing, and both foreign and domestic firms in these
cities experience rising productivity as well as rising profitability. However, the record
is more modest with regard to generating high-paying jobs for workers. No evidence is
found that average wage rates have increased in SEZs relative to other cities, in spite of
large increases in average productivity of firms in these areas. In fact, since the zones
have been characterized by rising prices, the welfare implications for residents of these
cities are ambiguous. While the SEZs have provided the opportunity for millions of rural
workers to find urban employment, these areas have not yet experienced real wage
growth.
As the PRC’s economy matures, it stands to reason that workers will increase their
bargaining power, and that increases in productivity will induce real wage increases. In
the short run, however, policy makers should consider measures that prevent workers
22 | ADB Economics Working Paper Series No. 276
from poor rural areas having a “race to the bottom”, where firms can pay individuals
subsistence wages and capture the lion’s share of the benefits to trade. Inequality within
the PRC’s cities is also a potential trigger for social unrest. Recent riots by low-wage
workers suggest that the status quo is problematic, as workers complain of unscrupulous
managers and corrupt local officials.7 Future policy should be designed with the strategy
of promoting growth but also providing a framework for establishing standards of worker
compensation and protection that will generate robust growth and a fairer distribution of
the benefits of trade.
7
See edition.cnn.com/2011/WORLD/asiapcf/06/17/china.riot.town.yoon/index.html?hpt=hp_t2.
Winners and Losers of Multinational Firm Entry into Developing Countries:
Evidence from the Special Economic Zones of the People’s Republic of China | 23
Appendix
Appendix Figure 1: Firm Openings in the PRC’s Special Economic Zones
Number of Firm Openings
3,000
2,000
1,000
0
1980
82
84
86
88
90
92
Domestic Firms
94
96
98
2000
Foreign Firms
Source: Author’s calculations.
Appendix Figure 2: Trends in Nominal Wages: Special Economic Zones and All Other Cities
40,000
Wages (Yuan)
30,000
20,000
10,000
0
1988
90
92
94
96
98
2000
Special Economic Zones (90th percentile)
Special Economic Zones (mean wage)
All Other Cities (90th percentile)
All Other Cities (mean wage)
Source: Author’s calculations.
24 | ADB Economics Working Paper Series No. 276
Appendix Figure 3: Trends in Real Wages: Special Economic Zones and All Other Cities
Real Wages (Yuan)
4,000
3,000
2,000
1,000
1988
90
92
94
96
98
2000
Special Economic Zones (median)
All Other Cities (median)
Source: Author’s calculations.
Appendix Figure 4: Firm Openings in Special Economic Zones
Number of Firm Openings
500
400
300
200
100
0
1980 82
84
86
88
90
Domestic Firms
Source: Author’s calculations.
92
94
96
98 2000
Foreign Firms
Winners and Losers of Multinational Firm Entry into Developing Countries:
Evidence from the Special Economic Zones of the People’s Republic of China | 25
Number of Domestic Firm Openings
Appendix Figure 5: Domestic Firms Openings over the Years
2,000
8000
1,500
6000
1,000
4000
500
2000
0
1980 82
84
86
88
Free Trade Zones
Coastal Cities
90
92
94
96
0
98 2000
Special Economic Zones
Other Areas (right axis)
Source: Author’s calculations.
References
Antras, P., L. Garicano, and E. Rossi-Hansberg. 2006. “Offshoring in a Knowledge Economy.”
Quarterly Journal of Economics 121(1):31–77.
Demurger, S., J. D. Sachs, W. T. Woo, S. Bao, and G. Chang. 2002. “The Relative Contributions
of Location and Preferential Policies in [People’s Republic of] China's Regional Development:
Being in the Right Place and Having the Right Incentives.” China Economic Review 13:444–
65.
Deng, F., and Y. Q. Huang. 2004. “Uneven Land Reform and Urban Sprawl in [People’s Republic
of] China: The Case of Beijing.” Progress in Planning 61:211–36.
Ebenstein, A., L. Du, A. Harrison, and G. Jefferson. 2011. Do Institutions Matter for FDI Spillovers?
The Implications of [People’s Republic of] China’s “Special Characteristics”. Policy Research
Working Paper 5757, The World Bank, Washington, DC.
Ebenstein, A., A. Harrison, M. McMillan, and S. Phillips. 2009. Estimating the Impact of Trade and
Offshoring Using the Current Population Surveys. NBER Working Paper 15107, National
Bureau of Economic Research, Massachusetts.
Grossman, G. M., and E. Rossi-Hansberg. 2009. “Trading Tasks: A Simple Theory of Offshoring.”
American Economic Review 98(5):1978–97.
Helpman, E., M. Melitz, and S. Yeaple. 2004. “Export Versus FDI with Heterogeneous Firms.”
American Economic Review 94(2):300–16.
Hsieh, C-T., and P. Klenow. 2009. “Misallocation and Manufacturing TFP in [People’s Republic of]
China and India.” Quarterly Journal of Economics 124(4):1403–48.
Jones, D. C., C. Li, and A. L. Owen. 2003. “Growth and Regional Inequality in [People’s Republic
of] China during the Reform Era.” China Economic Review 14:186–200.
Melitz, M. 2003. “The Impact of Trade on Intra-industry Reallocations and Aggregate Industry
Productivity.” Econometrica 71:1696–725.
Tyler, P. E. 1995. “Deng's Economic Drive Leaves Vast Regions of [People’s Republic of] China
Behind.” The New York Times, 27 December. Available: www.nytimes.com/1995/12/27/world/
26 | ADB Economics Working Paper Series No. 276
deng-s-economic-drive-leaves-vast-regions-of-china-behind.html?pagewanted=all&src=pm.
Downloaded 5 May 2011.
Wong, S-W., and B. S. Tang. 2005. “Challenges to the Sustainability of ‘Development Zones’:
A Case Study of Guangzhou Development District, [People’s Republic of] China.” Cities
22(4):303–16.
Yeung, Y., J. Lee, and G. Kee. 2009. “[People’s Republic of] China’s Special Economic Zones at
30.” Eurasian Geography and Economics 50(2):222–40.
Zhang, J. 2011. “Interjurisdictional Competition for FDI: The Case of [People’s Republic of] China's
‘Development Zone Fever’.” Regional Science and Urban Economics 41:145–59.
About the Paper
Avraham Ebenstein examines the success of the People’s Republic of China’s special
economic zones (SEZs) at attracting high-productivity firms and whether this has led to
higher wages for workers. The paper finds that the SEZ experiment effectively attracted
foreign capital and high-productivity domestic firms. However, the results indicate
modest real wage gains for workers, suggesting firms have captured majority of benefits
of SEZs through higher profits.
About the Asian Development Bank
ADB’s vision is an Asia and Pacific region free of poverty. Its mission is to help its
developing member countries reduce poverty and improve the quality of life of their
people. Despite the region’s many successes, it remains home to two-thirds of the
world’s poor: 1.8 billion people who live on less than $2 a day, with 903 million
struggling on less than $1.25 a day. ADB is committed to reducing poverty through
inclusive economic growth, environmentally sustainable growth, and regional
integration.
Based in Manila, ADB is owned by 67 members, including 48 from the region. Its
main instruments for helping its developing member countries are policy dialogue, loans,
equity investments, guarantees, grants, and technical assistance.
Asian Development Bank
6 ADB Avenue, Mandaluyong City
1550 Metro Manila, Philippines
www.adb.org/economics
ISSN: 1655-5252
Publication Stock No. WPS114146
October 2011
< 0114 1467 >
Printed on recycled paper
Printed in the Philippines